On 2016/04/16 3:39 PM, Cecil Westerhof wrote: > 2016-04-16 14:52 GMT+02:00 R Smith <rsmith at rsweb.co.za>: > >> -- 2016-04-16 14:44:55.054 | [Success] Script Success. >> >> As you can see, the INSERT obviously takes some time (even more-so if the >> CHECK constraint is added), but the DROP Table takes almost no time here... >> > ?The drop is a very big difference: .7 seconds or 13. Is almost 20 times as > long. Could I be doing something wrong??
The big one is still running, so we'll see - but I do think already there is something very different. Simon might be on to something with the cache. The journal mode should also make a difference, but not that big I think. Once this run finishes, I will try the 10-Mil one with different journal modes. I have never noticed a large time taken for dropping tables though, but then I do not often drop very large tables. > If useful, I could share the program I am using. As long as the version is known and the modes used, the rest of the software differences should be small - it's usually down to hardware.